This products liability action stems from a tire explosion and resulting automobile wreck in which one person was killed and two people were injured. The jury found for the defendant tire manufacturer, the Court of Appeals reversed and remanded for retrial,1 and this Court granted certiorari. Certiorari was not granted to address the grounds of the Court of Appeals’ reversal (hence, this matter will be retried),2 but rather was granted in order to address the Court of Appeals’ ruling on an evidentiary issue that it deemed likely to recur on retrial.

We now conclude that the Court of Appeals erred by ruling that the trial court was required to admit evidence of consumer claims honored by the tire manufacturer for tires manufactured at the tire plant where the tire involved in this action was manufactured. Without an independent showing of a substantial similarity between the purported tire defect that caused the injuries in this case and the basis for the consumer claims that were honored by the manufacturer, this evidence was inadmissible. Therefore, we reverse that portion of the Court of Appeals’ ruling.

In June 1991, Bobby Crosby was driving his wife Jan and his daughter Kelly to his parents’ home when the left rear tire of his Bronco II blew out, causing the vehicle to flip over. Bobby Crosby was killed, and Jan and Kelly Crosby were seriously injured. Acting on behalf of herself, as administratrix of Bobby’s estate, and as next friend of Kelly, Jan Crosby brought suit against the manufacturer of the left rear tire, Cooper Tire & Rubber Company. Crosby’s complaint alleged negligent design, negligent manufacture, failure to warn, violation of the implied warranty of merchantability, and strict liability. The complaint claimed that the wreck was caused by a separation of the tire’s radial belting as a result of defective manufacturing. Cooper Tire countered that the subject tire had been driven for more than 30,000 miles, and that the blowout was caused by an impact with a road hazard.

At trial, Crosby sought to introduce evidence of Cooper Tire’s “adjustment statistics” — i.e., honored consumer claims — for all tires manufactured at the Cooper Tire Texarkana, Arkansas, tire plant where the purportedly defective tire in this matter was manu*455factored. Crosby sought to introduce these adjustment statistics for the nine-year period preceding her accident. Cooper Tire opposed admission of the adjustment statistics, arguing that there was no evidence of a substantial similarity between the tires reflected in the adjustment data and the tire at issue in Crosby’s lawsuit. In response, Crosby argued that the adjustment data was admissible without a showing of substantial similarity.3

After considering these arguments, the trial court excluded the proffered adjustment statistics, concluding that the evidence was “generally inadmissible.” In November 1997, the case was tried before a jury, and a defendant’s verdict was promptly returned.

The Court of Appeals reversed and remanded, and in so doing held that the trial court erred in excluding Cooper Tire’s adjustment statistics. This Court granted certiorari, and, for the reasons explained below, we now reverse.

1. The Court of Appeals erred by disregarding the trial court’s correct conclusion that the tire adjustment data was inadmissible because Crosby failed to show a substantial similarity between the belt separation alleged to have caused the car wreck in this matter, and the reasons why the tires reflected in the adjustment data were returned by other consumers.

In products liability cases, the “rule of substantial similarity” prohibits the admission into evidence of other transactions, occurrences, or claims unless the proponent first shows that there is a “substantial similarity” between the other transactions, occurrences, or claims and the claim at issue in the litigation.4 The showing of substantial similarity must include a showing of similarity as to causation.5 Before admitting proffered evidence of other transactions in products liability cases, the trial court must satisfy itself that the rule of substantial similarity has been met.6 Thus, in Uniroyal Goodrich Tire Co. v. Ford, the Court of Appeals has recently held that it was reversible error to admit tire adjustment data in a products liability case without a “showing of similarity of the tires, defects or the causes thereof.”7

In this case, Crosby sought to introduce nine years’ worth of adjustment data concerning all types of tires manufactured at Cooper Tire’s Texarkana manufacturing plant. Crosby made no *456showing before the trial court that all of the tires reflected in the adjustment data were the same make and model as the tire involved in the accident at issue. Nor did Crosby show the trial court that the tires reflected in the adjustment data suffered from the same defect as the defect alleged in her suit. Finally, Crosby did not show similarity as to causation between the purported defect in her tire and any defects alleged in the adjusted tires.

Thus, the adjustment evidence was proffered at trial without any showing that: (1) Crosby’s tire and the adjusted tires shared a common design and manufacturing process; (2) suffered from a common defect; and (3) that any common defects shared the same causation. Without these showings, it was impossible for the rule of substantial similarity to be satisfied in this matter.8 Therefore, the trial court did not abuse its discretion in excluding the adjustment data evidence from trial, and the Court of Appeals erred in ruling otherwise.

2. The Court of Appeals also erred by holding that the rule of substantial similarity:

[D]oes not mean that the defects, tires, or occurrences must be identical, but only sufficiently substantially similar to be probative so that a jury can reasonably draw an inference of defect, causation, dangerousness, knowledge, producing the tire failure, or failure to warn. . . . [T]his means [there is] a continuum of admissible substantially similar occurrences, where at one extreme the occurrences were the same and at the other extreme the occurrences were barely sufficient to be substantially similar for admission so as not to be an abuse of discretion.9

We find no precedential directive from this Court (or any other court) to indicate that the admissibility of prior occurrences in products liability cases rests upon a continuum so wide as to include occurrences that are “identical” to the occurrence at issue as well as occurrences that are “barely” similar to the occurrence at issue. Under this extremely broad standard, almost all prior occurrence evidence would be admissible in products liability cases, so long as there was the barest hint of possible similarity between the prior occurrence and the occurrence at issue in litigation. In that situation, a trial court’s exclusion of such evidence would almost always be subject to a claim of discretionary abuse by a disgruntled litigant.

This Court has made it quite clear that the admission of evi*457dence is generally committed to the sound discretion of the trial court, “whose determination shall not be disturbed on appeal unless it amounts to an abuse of discretion.”10 This is so because trial courts, unlike appellate courts, are familiar with a piece of litigation from its inception, hear first-hand the arguments of counsel, and consider disputed evidence within the context of an entire proceeding. Hence, it is only natural that an appellate court should defer to the trial court with regard to the admission of evidence, unless the lower court’s decision is so flawed as to constitute an abuse of discretion.11 The Court of Appeals’ ruling quoted above is essentially a renunciation of the rule of substantial similarity enunciated in Mack Trucks, discussed supra, and its progeny. While purporting to give some deference to the trial court’s decision regarding the admissibility of similar occurrences, the Court of Appeals’ ruling is so erroneously all-inclusive as to essentially strip the trial court of its discretion to exclude evidence of similar occurrences in products liability cases when such evidence does not satisfy the rule of substantial similarity. Absent clear abuse, the trial courts’ exercise of discretion in admitting or refusing to admit such evidence is entitled to deference, and should not be hamstrung by restrictive rulings. Accordingly, the Court of Appeals’ ruling in this matter is hereby reversed.12

Judgment reversed.

dissenting.

Jan Crosby appealed a defense verdict in favor of Cooper Tire & Rubber Company to the Court of Appeals, which reversed the judgment on a ground not before us on writ of certiorari. Crosby v. Cooper Tire & Rubber Co., 240 Ga. App. 857 (1) (524 SE2d 313) (1999). That *458court then considered those other enumerations of error raised by Crosby which “are likely to again arise on retrial,” id. at 859 (1), noting that “many of the issues are based upon the sound exercise of the trial court’s discretion, which may change upon retrial before another trial judge. [Cit.]” Id. The adjustment evidence proffered by Crosby and excluded by the trial court is one of those issues which the Court of Appeals recognized could recur and its review of that evidence was tailored in light of the probability that the evidence would be proffered and ruled upon again. Based on my review of the Court of Appeals’ opinion, I find no error in its rulings or the language it used.

The trial court excluded the adjustment evidence proffered by Crosby finding it to be generally inadmissible. Compare Ray v. Ford Motor Co., 237 Ga. App. 316, 317 (1) (514 SE2d 227) (1999) (trial court excluded evidence “stating that Ray had failed to provide enough information to allow the court to determine whether incidents in the database were substantially similar to Ray’s accident”). In addressing the exclusion of this evidence, the Court of Appeals initially recognized that a material issue of fact in Crosby’s suit was whether the radial belting on the Bronco’s left rear tire separated because of a manufacturing defect or because the tire had 30,000 to 40,000 miles of wear on it. The Court of Appeals first found that upon retrial, Crosby’s adjustment evidence would be relevant and material, subject to the giving of limiting instructions, to show that tires fail for different reasons, some of which are related to consumer abuse with others related to manufacturing defects, as well as to show the frequency of manufacturing defects occurring after 30,000 to 40,000 miles of use. Id. at (2) (a). The court next found relevant and material upon retrial those adjustment documents, involving the same model tires from the same plant with prior or subsequent serial numbers and with “substantially similar manufacturing defects,” id. at (2) (b), “or that such model tire had whatever incidence of manufacturing defects as compared with overall production of such tire,” citing Browning v. Paccar, Inc., 214 Ga. App. 496 (1) (a) (448 SE2d 260) (1994) and other cases which have recognized the relevancy of such circumstantial evidence. Finally, the Court of Appeals applied this Court’s holding in Mack Trucks v. Conkle, 263 Ga. 539 (3) (436 SE2d 635) (1993) to hold that upon retrial, evidence of prior similar manufacturing defects would be admissible “to show defects, notice, prior knowledge and causation and for purposes of punitive damages, [cits.]” id. at (2) (c) but cautioned that for the admission of evidence on these bases, a foundation must be laid “that such occurrences were with the same or similar model steel belted tires, made with the same or similar materials and manufacturing processes, and having substantially similar defects.” Id.

I see no error in the Court of Appeals’ holdings in regard to the *459considerations applicable to the admission by the trial court of the adjustment evidence Crosby may choose to proffer upon retrial of this case. It is certainly important for a jury to know that Cooper Tire’s products have failed for many different reasons, including reasons assigned by both the plaintiff and the defendant for the blow-out of the Bronco’s left rear tire. The Court of Appeals’ opinion makes it clear that evidence admitted for this purpose must be accompanied by limiting instructions so that the jury does not consider it for any improper purpose. “The Georgia rule favors admissibility. If the relevancy of the offered evidence is in doubt, it should be admitted and sent to the jury under proper instructions.” McEachern v. McEachern, 260 Ga. 320, 321 (394 SE2d 92) (1990). This evidentiary rule favoring the admission of any relevant evidence is equally applicable in product liability cases. See, e.g., Wilson Foods Corp. v. Turner, 218 Ga. App. 74 (1) (460 SE2d 532) (1995) (applying “the evidentiary rule in Georgia favoring the admission of any relevant evidence, no matter how slight its probative value”). As to the admission of adjustment evidence to show the incidence of manufacturing defects as compared with overall production of such tire, Crosby, supra at (2) (b), it is well-established that circumstantial evidence relevant to prove a manufacturing defect may include evidence of the existence of the defect in goods produced at the same plant at around the same time. See Rose v. Figgie Intl., 229 Ga. App. 848 (1) (b) (495 SE2d 77) (1997); Browning v. Paccar, Inc., supra, 214 Ga. App. at 497 (1) (a); Carsten v. Wilkes Supermarket, 181 Ga. App. 834 (1) (353 SE2d 922) (1987). See also Central of Ga. R. Co. v. Bernstein, 113 Ga. 175 (1) (38 SE 394) (1901). This ruling relates to the admission of the adjustment evidence for purposes of showing manufacturing defects in other tires, regarding which the Court of Appeals’ opinion clearly reiterates the “substantial similarity” rule and holds that it is the standard which should be applied by the trial court in determining whether to admit the evidence upon retrial of this case.

The majority is not satisfied that the Court of Appeals explicitly states that the “substantial similarity” standard applies upon retrial to determine the admissibility of the proffered evidence. Instead, the majority concludes that the lower court “renounced” the standard it expressly utilized based on language in the Court of Appeals’ opinion about the “continuum of admissible substantially similar occurrences.” The majority’s criticism of this dicta is not justified. The Court of Appeals with this colorful language was simply recognizing what every experienced trial judge knows, namely, that there are no bright-line, hard-and-fast rules applicable to every case when it comes to the admissibility of evidence of similar occurrences. What constitutes “substantially similar” evidence does indeed shift between the extremes of absolutely identical to “barely sufficient to *460be substantially similar,” with all the gradations in-between. There is no exact science involved in weighing such evidence: evidence that a trial court determines is substantially similar in one tire blow-out case may be found inadequate by another trial court in another tire blow-out case. It is for this very reason that we have consistently applied a “manifest abuse” standard in assessing a trial court’s discretion in the admission or exclusion of evidence.

The Court of Appeals has correctly recognized that in determining the admissibility of similar occurrences, a trial court cannot abuse its discretion when the evidence it admits is sufficiently “probative so that a jury can reasonably draw an inference of defect, causation, dangerousness, knowledge, producing the tire failure, or failure to warn. [Cits.]” Crosby, supra at (2) (c). Accord Green, Georgia Law of Evidence (5th ed.), § 68, p. 147 (the rule regarding other transactions “is that evidence of other transactions or occurrences is admissible if it is relevant in the particular instance and does not present too great a danger of undue consumption of time, confusion of issues, undue prejudice or unfair surprise”). This recognition does not constitute a renunciation of Mack Trucks, supra, but rather endorses the holding in Mack Trucks that substantially similar evidence is admissible because it is relevant to the issues of notice and punitive damages and evidence that is “wholly different” should be excluded. Id., 263 Ga. at 544 (3).

From the stress the majority places on the “substantial” nature of similar occurrences in products liability cases and the criticism heaped on the Court of Appeals for recognizing the leeway that exists in determining what may constitute similar evidence in such cases, it appears that the majority is creating a stricter standard for the admission of relevant similar transaction evidence in product liability cases than in other cases. Even in criminal cases, where a person’s life and liberty are at stake, the courts only require that there be a “sufficient connection or similarity between the independent offense or act and the crime charged so that proof of the former tends to prove the latter. [Cit.]” Williams v. State, 261 Ga. 640, 642 (2) (409 SE2d 649) (1991). See, e.g., Quezada v. State, 236 Ga. App. 718, 720 (1) (512 SE2d 401) (1999) (appellate court need look only to whether similar transaction evidence in criminal trial was “sufficiently similar”). I cannot hold civil plaintiffs to a higher standard of proof to recover damages from the manufacturers of defective products than is required of the State to establish a defendant’s guilt beyond a reasonable doubt under constitutional standards.

I object to the majority’s opinion because it establishes an unwarranted threshold standard for admissibility in products liability cases that is contrary to the Georgia rule favoring the admission of any relevant evidence. The majority misreads the Court of *461Appeals’ opinion, in which that court sought only to address problems that may recur when the adjustment evidence is proffered upon retrial, and ignores the fact that the trial court here never expressly assessed the substantial similarity of the proffered evidence before excluding it. Most egregiously, the majority does exactly what it accuses the Court of Appeals of doing: limiting the discretion of trial courts. To paraphrase the majority opinion, its ruling is so erroneously under-inclusive as to essentially strip trial courts of their discretion to admit evidence of similar occurrences in products liability cases when such evidence fails to satisfy the untenably rigorous standard the majority now imposes. Because I find that the Court of Appeals set forth the proper standard for the trial court to apply should Crosby’s evidence be proffered again upon retrial, I would affirm that court’s rulings. Accordingly, I must respectfully dissent.

I am authorized to state that Chief Justice Benham joins in this dissent.